All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace
@ 2013-12-16  8:12 Atzm Watanabe
  2013-12-16 10:18 ` Daniel Borkmann
  2013-12-16 17:25 ` Stephen Hemminger
  0 siblings, 2 replies; 5+ messages in thread
From: Atzm Watanabe @ 2013-12-16  8:12 UTC (permalink / raw)
  To: netdev
  Cc: Stephen Hemminger, Ben Hutchings, David Miller, Daniel Borkmann,
	David Laight

This enables userspace to get VLAN TPID as well as the VLAN TCI.

Signed-off-by: Atzm Watanabe <atzm@stratosphere.co.jp>
---
 include/uapi/linux/if_packet.h | 23 +++++++++++++----------
 net/packet/af_packet.c         | 13 +++++++++----
 2 files changed, 22 insertions(+), 14 deletions(-)

diff --git a/include/uapi/linux/if_packet.h b/include/uapi/linux/if_packet.h
index 9185dc9..22d5192 100644
--- a/include/uapi/linux/if_packet.h
+++ b/include/uapi/linux/if_packet.h
@@ -84,17 +84,18 @@ struct tpacket_auxdata {
 	__u16		tp_mac;
 	__u16		tp_net;
 	__u16		tp_vlan_tci;
-	__u16		tp_padding;
+	__u16		tp_vlan_tpid;
 };
 
 /* Rx ring - header status */
-#define TP_STATUS_KERNEL	      0
-#define TP_STATUS_USER		(1 << 0)
-#define TP_STATUS_COPY		(1 << 1)
-#define TP_STATUS_LOSING	(1 << 2)
-#define TP_STATUS_CSUMNOTREADY	(1 << 3)
-#define TP_STATUS_VLAN_VALID	(1 << 4) /* auxdata has valid tp_vlan_tci */
-#define TP_STATUS_BLK_TMO	(1 << 5)
+#define TP_STATUS_KERNEL		      0
+#define TP_STATUS_USER			(1 << 0)
+#define TP_STATUS_COPY			(1 << 1)
+#define TP_STATUS_LOSING		(1 << 2)
+#define TP_STATUS_CSUMNOTREADY		(1 << 3)
+#define TP_STATUS_VLAN_VALID		(1 << 4) /* auxdata has valid tp_vlan_tci */
+#define TP_STATUS_BLK_TMO		(1 << 5)
+#define TP_STATUS_VLAN_TPID_VALID	(1 << 6) /* auxdata has valid tp_vlan_tpid */
 
 /* Tx ring - header status */
 #define TP_STATUS_AVAILABLE	      0
@@ -133,12 +134,14 @@ struct tpacket2_hdr {
 	__u32		tp_sec;
 	__u32		tp_nsec;
 	__u16		tp_vlan_tci;
-	__u8		tp_padding[6];
+	__u16		tp_vlan_tpid;
+	__u8		tp_padding[4];
 };
 
 struct tpacket_hdr_variant1 {
 	__u32	tp_rxhash;
 	__u32	tp_vlan_tci;
+	__u32	tp_vlan_tpid;
 };
 
 struct tpacket3_hdr {
@@ -154,7 +157,7 @@ struct tpacket3_hdr {
 	union {
 		struct tpacket_hdr_variant1 hv1;
 	};
-	__u8		tp_padding[12];
+	__u8		tp_padding[8];
 };
 
 struct tpacket_bd_ts {
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 5c75a1d..58d4fd7 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -977,9 +977,11 @@ static void prb_fill_vlan_info(struct tpacket_kbdq_core *pkc,
 {
 	if (vlan_tx_tag_present(pkc->skb)) {
 		ppd->hv1.tp_vlan_tci = vlan_tx_tag_get(pkc->skb);
-		ppd->tp_status = TP_STATUS_VLAN_VALID;
+		ppd->hv1.tp_vlan_tpid = (__force __u32)ntohs(pkc->skb->vlan_proto);
+		ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
 	} else {
 		ppd->hv1.tp_vlan_tci = 0;
+		ppd->hv1.tp_vlan_tpid = 0;
 		ppd->tp_status = TP_STATUS_AVAILABLE;
 	}
 }
@@ -1925,9 +1927,11 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		h.h2->tp_nsec = ts.tv_nsec;
 		if (vlan_tx_tag_present(skb)) {
 			h.h2->tp_vlan_tci = vlan_tx_tag_get(skb);
-			status |= TP_STATUS_VLAN_VALID;
+			h.h2->tp_vlan_tpid = ntohs(skb->vlan_proto);
+			status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
 		} else {
 			h.h2->tp_vlan_tci = 0;
+			h.h2->tp_vlan_tpid = 0;
 		}
 		hdrlen = sizeof(*h.h2);
 		memset(h.h2->tp_padding, 0,
@@ -2877,11 +2881,12 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock,
 		aux.tp_net = skb_network_offset(skb);
 		if (vlan_tx_tag_present(skb)) {
 			aux.tp_vlan_tci = vlan_tx_tag_get(skb);
-			aux.tp_status |= TP_STATUS_VLAN_VALID;
+			aux.tp_vlan_tpid = ntohs(skb->vlan_proto);
+			aux.tp_status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
 		} else {
 			aux.tp_vlan_tci = 0;
+			aux.tp_vlan_tpid = 0;
 		}
-		aux.tp_padding = 0;
 		put_cmsg(msg, SOL_PACKET, PACKET_AUXDATA, sizeof(aux), &aux);
 	}
 
-- 
1.8.1.5

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace
  2013-12-16  8:12 [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace Atzm Watanabe
@ 2013-12-16 10:18 ` Daniel Borkmann
  2013-12-16 12:41   ` Atzm Watanabe
  2013-12-16 17:25 ` Stephen Hemminger
  1 sibling, 1 reply; 5+ messages in thread
From: Daniel Borkmann @ 2013-12-16 10:18 UTC (permalink / raw)
  To: Atzm Watanabe
  Cc: netdev, Stephen Hemminger, Ben Hutchings, David Miller, David Laight

On 12/16/2013 09:12 AM, Atzm Watanabe wrote:
> This enables userspace to get VLAN TPID as well as the VLAN TCI.
>
> Signed-off-by: Atzm Watanabe <atzm@stratosphere.co.jp>

Looks good, one comment below though:

> ---
>   include/uapi/linux/if_packet.h | 23 +++++++++++++----------
>   net/packet/af_packet.c         | 13 +++++++++----
>   2 files changed, 22 insertions(+), 14 deletions(-)
>
> diff --git a/include/uapi/linux/if_packet.h b/include/uapi/linux/if_packet.h
> index 9185dc9..22d5192 100644
> --- a/include/uapi/linux/if_packet.h
> +++ b/include/uapi/linux/if_packet.h
> @@ -84,17 +84,18 @@ struct tpacket_auxdata {
>   	__u16		tp_mac;
>   	__u16		tp_net;
>   	__u16		tp_vlan_tci;
> -	__u16		tp_padding;
> +	__u16		tp_vlan_tpid;
>   };
>
>   /* Rx ring - header status */
> -#define TP_STATUS_KERNEL	      0
> -#define TP_STATUS_USER		(1 << 0)
> -#define TP_STATUS_COPY		(1 << 1)
> -#define TP_STATUS_LOSING	(1 << 2)
> -#define TP_STATUS_CSUMNOTREADY	(1 << 3)
> -#define TP_STATUS_VLAN_VALID	(1 << 4) /* auxdata has valid tp_vlan_tci */
> -#define TP_STATUS_BLK_TMO	(1 << 5)
> +#define TP_STATUS_KERNEL		      0
> +#define TP_STATUS_USER			(1 << 0)
> +#define TP_STATUS_COPY			(1 << 1)
> +#define TP_STATUS_LOSING		(1 << 2)
> +#define TP_STATUS_CSUMNOTREADY		(1 << 3)
> +#define TP_STATUS_VLAN_VALID		(1 << 4) /* auxdata has valid tp_vlan_tci */
> +#define TP_STATUS_BLK_TMO		(1 << 5)
> +#define TP_STATUS_VLAN_TPID_VALID	(1 << 6) /* auxdata has valid tp_vlan_tpid */
>
>   /* Tx ring - header status */
>   #define TP_STATUS_AVAILABLE	      0
> @@ -133,12 +134,14 @@ struct tpacket2_hdr {
>   	__u32		tp_sec;
>   	__u32		tp_nsec;
>   	__u16		tp_vlan_tci;
> -	__u8		tp_padding[6];
> +	__u16		tp_vlan_tpid;
> +	__u8		tp_padding[4];
>   };
>
>   struct tpacket_hdr_variant1 {
>   	__u32	tp_rxhash;
>   	__u32	tp_vlan_tci;
> +	__u32	tp_vlan_tpid;

Why is this not __u16? If this needs to be __u32 aligned, then
probably we should pad with another __u16. I think we should not
waste any space here in case there are future additions of new
members.

>   };
>
>   struct tpacket3_hdr {
> @@ -154,7 +157,7 @@ struct tpacket3_hdr {
>   	union {
>   		struct tpacket_hdr_variant1 hv1;
>   	};
> -	__u8		tp_padding[12];
> +	__u8		tp_padding[8];
>   };
>
>   struct tpacket_bd_ts {
> diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
> index 5c75a1d..58d4fd7 100644
> --- a/net/packet/af_packet.c
> +++ b/net/packet/af_packet.c
> @@ -977,9 +977,11 @@ static void prb_fill_vlan_info(struct tpacket_kbdq_core *pkc,
>   {
>   	if (vlan_tx_tag_present(pkc->skb)) {
>   		ppd->hv1.tp_vlan_tci = vlan_tx_tag_get(pkc->skb);
> -		ppd->tp_status = TP_STATUS_VLAN_VALID;
> +		ppd->hv1.tp_vlan_tpid = (__force __u32)ntohs(pkc->skb->vlan_proto);

Nit:                                             ^^^^ one space here, resp. with __u16
                                                       you won't need the cast.

> +		ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
>   	} else {
>   		ppd->hv1.tp_vlan_tci = 0;
> +		ppd->hv1.tp_vlan_tpid = 0;
>   		ppd->tp_status = TP_STATUS_AVAILABLE;
>   	}
>   }
> @@ -1925,9 +1927,11 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
>   		h.h2->tp_nsec = ts.tv_nsec;
>   		if (vlan_tx_tag_present(skb)) {
>   			h.h2->tp_vlan_tci = vlan_tx_tag_get(skb);
> -			status |= TP_STATUS_VLAN_VALID;
> +			h.h2->tp_vlan_tpid = ntohs(skb->vlan_proto);
> +			status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
>   		} else {
>   			h.h2->tp_vlan_tci = 0;
> +			h.h2->tp_vlan_tpid = 0;
>   		}
>   		hdrlen = sizeof(*h.h2);
>   		memset(h.h2->tp_padding, 0,
> @@ -2877,11 +2881,12 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock,
>   		aux.tp_net = skb_network_offset(skb);
>   		if (vlan_tx_tag_present(skb)) {
>   			aux.tp_vlan_tci = vlan_tx_tag_get(skb);
> -			aux.tp_status |= TP_STATUS_VLAN_VALID;
> +			aux.tp_vlan_tpid = ntohs(skb->vlan_proto);
> +			aux.tp_status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
>   		} else {
>   			aux.tp_vlan_tci = 0;
> +			aux.tp_vlan_tpid = 0;
>   		}
> -		aux.tp_padding = 0;
>   		put_cmsg(msg, SOL_PACKET, PACKET_AUXDATA, sizeof(aux), &aux);
>   	}
>
>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace
  2013-12-16 10:18 ` Daniel Borkmann
@ 2013-12-16 12:41   ` Atzm Watanabe
  0 siblings, 0 replies; 5+ messages in thread
From: Atzm Watanabe @ 2013-12-16 12:41 UTC (permalink / raw)
  To: Daniel Borkmann
  Cc: netdev, Stephen Hemminger, Ben Hutchings, David Miller, David Laight

At Mon, 16 Dec 2013 11:18:09 +0100,
Daniel Borkmann wrote:
> 
> On 12/16/2013 09:12 AM, Atzm Watanabe wrote:
> > This enables userspace to get VLAN TPID as well as the VLAN TCI.
> >
> > Signed-off-by: Atzm Watanabe <atzm@stratosphere.co.jp>
> 
> Looks good, one comment below though:

Thank you for comments.


> >   	__u32	tp_rxhash;
> >   	__u32	tp_vlan_tci;
> > +	__u32	tp_vlan_tpid;
> 
> Why is this not __u16? If this needs to be __u32 aligned, then
> probably we should pad with another __u16. I think we should not
> waste any space here in case there are future additions of new
> members.

Ah, indeed.  I think it does not need to be __u32 aligned too.
I'll fix it next time.


> > diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
> > index 5c75a1d..58d4fd7 100644
> > --- a/net/packet/af_packet.c
> > +++ b/net/packet/af_packet.c
> > @@ -977,9 +977,11 @@ static void prb_fill_vlan_info(struct tpacket_kbdq_core *pkc,
> >   {
> >   	if (vlan_tx_tag_present(pkc->skb)) {
> >   		ppd->hv1.tp_vlan_tci = vlan_tx_tag_get(pkc->skb);
> > -		ppd->tp_status = TP_STATUS_VLAN_VALID;
> > +		ppd->hv1.tp_vlan_tpid = (__force __u32)ntohs(pkc->skb->vlan_proto);
> 
> Nit:                                             ^^^^ one space here, resp. with __u16
>                                                        you won't need the cast.

Yes, will remove the cast next time.


Thank you again!

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace
  2013-12-16  8:12 [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace Atzm Watanabe
  2013-12-16 10:18 ` Daniel Borkmann
@ 2013-12-16 17:25 ` Stephen Hemminger
  2013-12-17  5:51   ` Atzm Watanabe
  1 sibling, 1 reply; 5+ messages in thread
From: Stephen Hemminger @ 2013-12-16 17:25 UTC (permalink / raw)
  To: Atzm Watanabe
  Cc: netdev, Ben Hutchings, David Miller, Daniel Borkmann, David Laight

On Mon, 16 Dec 2013 17:12:09 +0900
Atzm Watanabe <atzm@stratosphere.co.jp> wrote:

>  struct tpacket_hdr_variant1 {
>  	__u32	tp_rxhash;
>  	__u32	tp_vlan_tci;
> +	__u32	tp_vlan_tpid;
>  };

Is this safe? Won't you break ABI for existing userspace applications.

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace
  2013-12-16 17:25 ` Stephen Hemminger
@ 2013-12-17  5:51   ` Atzm Watanabe
  0 siblings, 0 replies; 5+ messages in thread
From: Atzm Watanabe @ 2013-12-17  5:51 UTC (permalink / raw)
  To: Stephen Hemminger
  Cc: netdev, Ben Hutchings, David Miller, Daniel Borkmann, David Laight

At Mon, 16 Dec 2013 09:25:09 -0800,
Stephen Hemminger wrote:
> 
> On Mon, 16 Dec 2013 17:12:09 +0900
> Atzm Watanabe <atzm@stratosphere.co.jp> wrote:
> 
> >  struct tpacket_hdr_variant1 {
> >  	__u32	tp_rxhash;
> >  	__u32	tp_vlan_tci;
> > +	__u32	tp_vlan_tpid;
> >  };
> 
> Is this safe? Won't you break ABI for existing userspace applications.

I think we can do it safely without breaking userspace.
struct tpacket_hdr_variant1 is an union member of struct tpacket3_hdr,
and it is not used for any others in the kernel-userspace ABI.
struct tpacket3_hdr is aligned to a multiple of 16 bytes, its size is
currently 48 bytes.  Against that, the size before aligning is
currently 36 bytes, so there is a gap of 12 bytes that is unused.

Previous patch 2/3 adds padding members for filling the gap, after the
patch 1/3 which adds BUILD_BUG_ON() as sentinel for aligned size.
The patch 3/3 adds tp_vlan_tpid by picking area from the padding added
in the patch 2/3.

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2013-12-17  5:51 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-12-16  8:12 [PATCH v3 3/3] packet: Deliver VLAN TPID to userspace Atzm Watanabe
2013-12-16 10:18 ` Daniel Borkmann
2013-12-16 12:41   ` Atzm Watanabe
2013-12-16 17:25 ` Stephen Hemminger
2013-12-17  5:51   ` Atzm Watanabe

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.